Land Unit Movement

[hueglin]

Does anyone know:

1)do all land units have the same basic rate of movement?
2)what effect (in game terms) does making a device a 'vehicle type' have?
3)what effect (in game terms) does making a device a 'AFV type' have?
4)what effect (in game terms) does making a device a 'squad type' have?

I am trying to model cavalry vs infantry in the 1914 time period. My understanding of the differences being:
Cavalry (and Mounted Rifles) would have:
- less firepower (need horse holders if dismounted and firing)
- more tactical mobility (better for recce - if properly trained)
- similar operational mobility (roughly same rate of advance over days and weeks)
- less strategic mobility (if transported by rail or sea - take up more space)

I am thinking there is no real way to model cavalry's advantages in this game system. Any thoughts on this subject would be welcome?

[treespider]

ORIGINAL: hueglin

Does anyone know:

1)do all land units have the same basic rate of movement?
2)what effect (in game terms) does making a device a 'vehicle type' have?
3)what effect (in game terms) does making a device a 'AFV type' have?
4)what effect (in game terms) does making a device a 'squad type' have?

I am trying to model cavalry vs infantry in the 1914 time period. My understanding of the differences being:
Cavalry (and Mounted Rifles) would have:
- less firepower (need horse holders if dismounted and firing)
- more tactical mobility (better for recce - if properly trained)
- similar operational mobility (roughly same rate of advance over days and weeks)
- less strategic mobility (if transported by rail or sea - take up more space)

I am thinking there is no real way to model cavalry's advantages in this game system. Any thoughts on this subject would be welcome?

The manual has the movement rate tables by unit type in it somewhere...In addition the 'squad' type devices need a particular firapower value ..iirc 12 to register as a squad for AV purposes.

And yes you are probably correct in assuming that there are no apparent differences between cavalry and infantry in the game.

[el cid again]

Actually, cavalry does not have greater operational mobility. "Tactical" would be well within the hex on our scale.
Infantry could - and often did - outmarch cavalry over any distance. See in particular the Nez Pierce war (where true infantry and artillery acting as infantry sans cannon routinely outmarched even very good cavalry). Also see the French Foreign Legion - which doctrinally requires infantry to march 50 miles a day AND build a fort at the end of the day -
a feat cavalry probably never has achieved.

[hueglin]

Thanks, I didn't think of looking in the game manual - I just looked in the editor manual.

I agree that Cavalry (European) has similar operational mobility to infantry - there are some very interesting studies about rates of advance in combat by various types of forces (From Numbers, Predictions and War by Dupuy: Napoleon's Grande Armee averaged about 13 km a day advancing to Moscow vs the Wehrmacht rate of 7.5km a day. Also, the British/Commonwealth advance is Palestine in 1918 was achieved at roughly the same rate of advance as the Israeli advance in the Sinai during the Six Day War)

One cavalry force that did achieve high operational mobility was the Mongols. I seem to remember reading once that their rates of advance were similar to or better than mechanized units. European Cavalry is another matter entirely though.

[herwin]

ORIGINAL: el cid again

Actually, cavalry does not have greater operational mobility. "Tactical" would be well within the hex on our scale.
Infantry could - and often did - outmarch cavalry over any distance. See in particular the Nez Pierce war (where true infantry and artillery acting as infantry sans cannon routinely outmarched even very good cavalry). Also see the French Foreign Legion - which doctrinally requires infantry to march 50 miles a day AND build a fort at the end of the day -
a feat cavalry probably never has achieved.

Cavalry was able to move operationally without a supply line, and tactically it could move faster than infantry. That allowed it to screen and force infantry to deploy, an advantage that goes completely over the head of the game designers.

[herwin]

ORIGINAL: hueglin

Thanks, I didn't think of looking in the game manual - I just looked in the editor manual.

I agree that Cavalry (European) has similar operational mobility to infantry - there are some very interesting studies about rates of advance in combat by various types of forces (From Numbers, Predictions and War by Dupuy: Napoleon's Grande Armee averaged about 13 km a day advancing to Moscow vs the Wehrmacht rate of 7.5km a day. Also, the British/Commonwealth advance is Palestine in 1918 was achieved at roughly the same rate of advance as the Israeli advance in the Sinai during the Six Day War)

One cavalry force that did achieve high operational mobility was the Mongols. I seem to remember reading once that their rates of advance were similar to or better than mechanized units. European Cavalry is another matter entirely though.

The Mongols brought remounts.

By the way, Dupuy is incredibly bad. For one thing, his regressions had more parameters than data points. You need at least three sample points per parameter estimated--take a look at the number of battles in his database. I did similar work, but with a database of over 600 one-day battles in WWII. The resulting CRT is available here.

[hueglin]

ORIGINAL: herwin

ORIGINAL: hueglin

Thanks, I didn't think of looking in the game manual - I just looked in the editor manual.

I agree that Cavalry (European) has similar operational mobility to infantry - there are some very interesting studies about rates of advance in combat by various types of forces (From Numbers, Predictions and War by Dupuy: Napoleon's Grande Armee averaged about 13 km a day advancing to Moscow vs the Wehrmacht rate of 7.5km a day. Also, the British/Commonwealth advance is Palestine in 1918 was achieved at roughly the same rate of advance as the Israeli advance in the Sinai during the Six Day War)

One cavalry force that did achieve high operational mobility was the Mongols. I seem to remember reading once that their rates of advance were similar to or better than mechanized units. European Cavalry is another matter entirely though.

The Mongols brought remounts.

By the way, Dupuy is incredibly bad. For one thing, his regressions had more parameters than data points. You need at least three sample points per parameter estimated--take a look at the number of battles in his database. I did similar work, but with a database of over 600 one-day battles in WWII. The resulting CRT is available here.

Thanks for the info. The CRT looks interesting and well thought out. Do you play games with it? As for Dupuy, I'm not much of a statistician (or mathematician for that matter). I'm interested in differing viewpoints. What is your opinion of the OLI (Operational Lethality Index Dupuy uses? Also, what do you think of his conclusions that German land forces were 20% more effective, man for man, than Allied?

[hueglin]

ORIGINAL: treespider

ORIGINAL: hueglin

Does anyone know:

1)do all land units have the same basic rate of movement?
2)what effect (in game terms) does making a device a 'vehicle type' have?
3)what effect (in game terms) does making a device a 'AFV type' have?
4)what effect (in game terms) does making a device a 'squad type' have?

I am trying to model cavalry vs infantry in the 1914 time period. My understanding of the differences being:
Cavalry (and Mounted Rifles) would have:
- less firepower (need horse holders if dismounted and firing)
- more tactical mobility (better for recce - if properly trained)
- similar operational mobility (roughly same rate of advance over days and weeks)
- less strategic mobility (if transported by rail or sea - take up more space)

I am thinking there is no real way to model cavalry's advantages in this game system. Any thoughts on this subject would be welcome?

The manual has the movement rate tables by unit type in it somewhere...In addition the 'squad' type devices need a particular firapower value ..iirc 12 to register as a squad for AV purposes.

And yes you are probably correct in assuming that there are no apparent differences between cavalry and infantry in the game.

Thanks for the info about the minimum value of 12 - what happens with a value lower than 12?
I was planning to have an Anti-soft value of 10 for a 10 man section with bolt-action rifles? Are you sure it won't work?

Maybe it would be better to make the basic element a 30 man platoon - 30 Anti-soft and 30 load factor. What do you think? Would there be other disadvantages to this? The other possibility is to just make the Anti-soft 12 and count it as either 10 men or 12 men.

[herwin]

ORIGINAL: hueglin

ORIGINAL: herwin

ORIGINAL: hueglin

Thanks, I didn't think of looking in the game manual - I just looked in the editor manual.

I agree that Cavalry (European) has similar operational mobility to infantry - there are some very interesting studies about rates of advance in combat by various types of forces (From Numbers, Predictions and War by Dupuy: Napoleon's Grande Armee averaged about 13 km a day advancing to Moscow vs the Wehrmacht rate of 7.5km a day. Also, the British/Commonwealth advance is Palestine in 1918 was achieved at roughly the same rate of advance as the Israeli advance in the Sinai during the Six Day War)

One cavalry force that did achieve high operational mobility was the Mongols. I seem to remember reading once that their rates of advance were similar to or better than mechanized units. European Cavalry is another matter entirely though.

The Mongols brought remounts.

By the way, Dupuy is incredibly bad. For one thing, his regressions had more parameters than data points. You need at least three sample points per parameter estimated--take a look at the number of battles in his database. I did similar work, but with a database of over 600 one-day battles in WWII. The resulting CRT is available here.

Thanks for the info. The CRT looks interesting and well thought out. Do you play games with it? As for Dupuy, I'm not much of a statistician (or mathematician for that matter). I'm interested in differing viewpoints. What is your opinion of the OLI (Operational Lethality Index Dupuy uses? Also, what do you think of his conclusions that German land forces were 20% more effective, man for man, than Allied?

He does not have the evidence for either claim. On the other hand, the effectiveness differences were real, but involved higher cohesion and resistance to disruption, not 'effectiveness'.

My son (Michael Erwin) uses the CRT in some of the games he develops.

[el cid again]

ORIGINAL: herwin

ORIGINAL: hueglin

Thanks, I didn't think of looking in the game manual - I just looked in the editor manual.

I agree that Cavalry (European) has similar operational mobility to infantry - there are some very interesting studies about rates of advance in combat by various types of forces (From Numbers, Predictions and War by Dupuy: Napoleon's Grande Armee averaged about 13 km a day advancing to Moscow vs the Wehrmacht rate of 7.5km a day. Also, the British/Commonwealth advance is Palestine in 1918 was achieved at roughly the same rate of advance as the Israeli advance in the Sinai during the Six Day War)

One cavalry force that did achieve high operational mobility was the Mongols. I seem to remember reading once that their rates of advance were similar to or better than mechanized units. European Cavalry is another matter entirely though.

The Mongols brought remounts.

By the way, Dupuy is incredibly bad. For one thing, his regressions had more parameters than data points. You need at least three sample points per parameter estimated--take a look at the number of battles in his database. I did similar work, but with a database of over 600 one-day battles in WWII. The resulting CRT is available here.

Quite right about the Mongols. Each soldier started with Tammerlain started with no less than 5 horses - probably the extream case - and most of them ATE three of the horses in route!

Dupuy and Dupuy are incredably good. They achieved the impossible - that is something I never would have believed was possibly except for the evidence. They created a model that works for all eras in all types of warfare. And it works with a sense of precision I would not have believed possible - not telling you who will win but the distance of advance, the amount of casualties, in detail. This can be backfitted to historical battles for validation. Granted it is expensive to get at the data (unless you have a peculiar sort of access - there are two ways in without paying the up front fee - which is astromical). Anyone who thinks they are incredably bad (or who does not know "he" is a "they") has not looked at this stuff. Even the most skeptical - and I probably qualify as one of those - would have to say "they did very well indeed."

[el cid again]

ORIGINAL: herwin

ORIGINAL: el cid again

Actually, cavalry does not have greater operational mobility. "Tactical" would be well within the hex on our scale.
Infantry could - and often did - outmarch cavalry over any distance. See in particular the Nez Pierce war (where true infantry and artillery acting as infantry sans cannon routinely outmarched even very good cavalry). Also see the French Foreign Legion - which doctrinally requires infantry to march 50 miles a day AND build a fort at the end of the day -
a feat cavalry probably never has achieved.

Cavalry was able to move operationally without a supply line, and tactically it could move faster than infantry. That allowed it to screen and force infantry to deploy, an advantage that goes completely over the head of the game designers.

This is nonsense. Only the Mongols achieved anything like significant movement "without a supply line" - and that was done exactly once (Tammerlain crossed the Himalaya's). In that exceptional case it was necessary to win the first battle - or perish. All significant military forces need a supply line. Game designers are not wrong to think so.

[el cid again]

ORIGINAL: herwin


[
Thanks for the info. The CRT looks interesting and well thought out. Do you play games with it? As for Dupuy, I'm not much of a statistician (or mathematician for that matter). I'm interested in differing viewpoints. What is your opinion of the OLI (Operational Lethality Index Dupuy uses? Also, what do you think of his conclusions that German land forces were 20% more effective, man for man, than Allied?

He does not have the evidence for either claim. On the other hand, the effectiveness differences were real, but involved higher cohesion and resistance to disruption, not 'effectiveness'.

My son (Michael Erwin) uses the CRT in some of the games he develops.
[/quote]

The evidence is decisive - and clear. It also shows up in a variety of ways. It is politically incorrect in the USA - and the US Army in particular - to accept these numbers. Cohesion and resistence to disruption are a fine basis to define effectiveness - so posing them as somehow antithical is a red herring. It may be interesting why they were more effective - but it does not change that they were. SLAM - who eventually becaome even more unpopular than the Dupuy's - was adament: tell the truth no matter who does not like it. "We can only learn from the truth."

[el cid again]

ORIGINAL: treespider



The manual has the movement rate tables by unit type in it somewhere...In addition the 'squad' type devices need a particular firapower value ..iirc 12 to register as a squad for AV purposes.

And yes you are probably correct in assuming that there are no apparent differences between cavalry and infantry in the game.

It appears this is incorrect. AV values seem to be very high even if squad values are 0 or 1. Over a year of attempting to create squads that "don't count" (for supply sinks) makes this very clear. No one yet has found a way. It is certain even a value below 2 is not enough to prevent a squad from registering for AV purposes.

What IS true is that certain types of weapons must have certain values to count for certain functions. Thus, for example, there is a minimum value to cause code to consider a device to be artillery. This turns out to be complex: there is not only a minimum firepower but a minimum effectiveness. There are probably other - equally undocumented - limits. You can see this when you set the values too low - the artillery option to bombard is then greyed out even when the enemy is in the hex (it is always greyed out if there is no enemy in the hex).

[herwin]

ORIGINAL: el cid again

Dupuy and Dupuy are incredably good. They achieved the impossible - that is something I never would have believed was possibly except for the evidence. They created a model that works for all eras in all types of warfare. And it works with a sense of precision I would not have believed possible - not telling you who will win but the distance of advance, the amount of casualties, in detail. This can be backfitted to historical battles for validation. Granted it is expensive to get at the data (unless you have a peculiar sort of access - there are two ways in without paying the up front fee - which is astromical). Anyone who thinks they are incredably bad (or who does not know "he" is a "they") has not looked at this stuff. Even the most skeptical - and I probably qualify as one of those - would have to say "they did very well indeed."

They pleased their customers in the Pentagon, yes, but not the professionals in the field. When you're doing statistical modelling, you need at least three data points for each parameter you estimate. This includes interactions and higher order terms. They don't have the data to fit a model of that size. *And* given the importance of interactions in combat, their model is too small.

[Dili]

So how can we know that -the artillery values, cavalry- El Cid? Can we use RHS values comparatively?

[herwin]

ORIGINAL: el cid again

ORIGINAL: herwin

ORIGINAL: el cid again

Actually, cavalry does not have greater operational mobility. "Tactical" would be well within the hex on our scale.
Infantry could - and often did - outmarch cavalry over any distance. See in particular the Nez Pierce war (where true infantry and artillery acting as infantry sans cannon routinely outmarched even very good cavalry). Also see the French Foreign Legion - which doctrinally requires infantry to march 50 miles a day AND build a fort at the end of the day -
a feat cavalry probably never has achieved.

Cavalry was able to move operationally without a supply line, and tactically it could move faster than infantry. That allowed it to screen and force infantry to deploy, an advantage that goes completely over the head of the game designers.

This is nonsense. Only the Mongols achieved anything like significant movement "without a supply line" - and that was done exactly once (Tammerlain crossed the Himalaya's). In that exceptional case it was necessary to win the first battle - or perish. All significant military forces need a supply line. Game designers are not wrong to think so.

The Red Army cavalry corps gave the Germans a lot of difficulty because they would abandon their supply lines when necessary and move cross country.

[herwin]

ORIGINAL: el cid again

ORIGINAL: herwin

Thanks for the info. The CRT looks interesting and well thought out. Do you play games with it? As for Dupuy, I'm not much of a statistician (or mathematician for that matter). I'm interested in differing viewpoints. What is your opinion of the OLI (Operational Lethality Index Dupuy uses? Also, what do you think of his conclusions that German land forces were 20% more effective, man for man, than Allied?

He does not have the evidence for either claim. On the other hand, the effectiveness differences were real, but involved higher cohesion and resistance to disruption, not 'effectiveness'.

My son (Michael Erwin) uses the CRT in some of the games he develops.

The evidence is decisive - and clear. It also shows up in a variety of ways. It is politically incorrect in the USA - and the US Army in particular - to accept these numbers. Cohesion and resistence to disruption are a fine basis to define effectiveness - so posing them as somehow antithical is a red herring. It may be interesting why they were more effective - but it does not change that they were. SLAM - who eventually becaome even more unpopular than the Dupuy's - was adament: tell the truth no matter who does not like it. "We can only learn from the truth."

WWII German units were usually more effective than Allied, but it wasn't because (or 'just because') they did their tasks better when at full strength on a proving ground; it was because losses, disruption, and confusion affected them less in sustained combat. On the other hand, the average Japanese unit was about equal in effectiveness to the average late war Allied unit; the problem early in the war was that the Allied units were green--understrength, poorly trained, led by donkeys, and outthought.

[el cid again]

ORIGINAL: herwin

ORIGINAL: el cid again

Dupuy and Dupuy are incredably good. They achieved the impossible - that is something I never would have believed was possibly except for the evidence. They created a model that works for all eras in all types of warfare. And it works with a sense of precision I would not have believed possible - not telling you who will win but the distance of advance, the amount of casualties, in detail. This can be backfitted to historical battles for validation. Granted it is expensive to get at the data (unless you have a peculiar sort of access - there are two ways in without paying the up front fee - which is astromical). Anyone who thinks they are incredably bad (or who does not know "he" is a "they") has not looked at this stuff. Even the most skeptical - and I probably qualify as one of those - would have to say "they did very well indeed."

They pleased their customers in the Pentagon, yes, but not the professionals in the field. When you're doing statistical modelling, you need at least three data points for each parameter you estimate. This includes interactions and higher order terms. They don't have the data to fit a model of that size. *And* given the importance of interactions in combat, their model is too small.

Actually there were a great many more than three. But the full data set costs (if memory serves) $10,000 to access - and the openly published material is just a sort of sampler. But one contractor has published the whole set - and I got a copy on CD rom. Even so, the Dupuy Institute continues to insist on the fee if you try to get even a little bit of the data.
Don't confuse what is in two books for the whole data set - or you will come to invalid conclusions.

[el cid again]

ORIGINAL: herwin

ORIGINAL: el cid again

ORIGINAL: herwin




He does not have the evidence for either claim. On the other hand, the effectiveness differences were real, but involved higher cohesion and resistance to disruption, not 'effectiveness'.

My son (Michael Erwin) uses the CRT in some of the games he develops.

The evidence is decisive - and clear. It also shows up in a variety of ways. It is politically incorrect in the USA - and the US Army in particular - to accept these numbers. Cohesion and resistence to disruption are a fine basis to define effectiveness - so posing them as somehow antithical is a red herring. It may be interesting why they were more effective - but it does not change that they were. SLAM - who eventually becaome even more unpopular than the Dupuy's - was adament: tell the truth no matter who does not like it. "We can only learn from the truth."

WWII German units were usually more effective than Allied, but it wasn't because (or 'just because') they did their tasks better when at full strength on a proving ground; it was because losses, disruption, and confusion affected them less in sustained combat. On the other hand, the average Japanese unit was about equal in effectiveness to the average late war Allied unit; the problem early in the war was that the Allied units were green--understrength, poorly trained, led by donkeys, and outthought.

According to both an American general of reputation (Stilwell) and a Russian one (Zhukov) the Japanese were "the best soldiers in the world" - and "if you combined them with American officers they would have been the most effective army in the world" according to the former. Not that late war conscripted Japanese were very good - but line troops were formidable - and undefeated as a body when the war ended (a fact of considerable concern to leaders on both sides).

My USMC historian mentor taught that the Germans were more effective because they had a better doctrine taught to better standards (that is, everyone knew it). But the US was much more ad hoc and (to German thinking) for that reason unpredictable. The Americans might have things not in the TO&E and might try any sort of trick - making it up as they went. The Germans had a real doctrine and it was good on the average - which matters for statistics. On the other hand, a German small unit lost cohesion if the leader (regarded as "father") was lost - while an American unit would have someone else take over (usually but not always the next senior - the next senior sometimes would ask another to do it - if he believed it was more likely to work well).

The idea that early US soldiers were poor is dead wrong. The long serving professionals when the war began were very good - and better than the conscripts led by reservests that followed (and won the war). The US had serious problems in weapons (e.g. a lack of anti-tank weapons) and doctrine - and attitude. The flip side of "the Japanese can't be any good" was "they are supermen" when they turned out not to be as expected. US units had moments of glory in those dark early days - the wholly green but superbly trained Philippine Scouts only gave ground to front line IJA units because the units on their flanks retreated. It isn't the fault of US troops their leader (Mac) was a fool - and led them into a swamp where 100% got malaria AND denge fever - which combined with starvation doomed them to be unfit. When the Japanese attack came in on Bataan - the US line broke in 30 minutes. The Japanese had used their trump (the ONLY time we ever faced it): heavy guns (IJA was "artillery happy" and believed these would work - and they did on Bataan). Had those same troops fallen back on Baguio City - in mountains - with multiple river lines to defend - and a rice growing area at their back - they not only would have denied to the enemy significant mines (including the largest copper mine in Asia) - they probably would have still been in the field when we returned to the area.

[herwin]

ORIGINAL: el cid again

ORIGINAL: herwin

ORIGINAL: el cid again

Dupuy and Dupuy are incredably good. They achieved the impossible - that is something I never would have believed was possibly except for the evidence. They created a model that works for all eras in all types of warfare. And it works with a sense of precision I would not have believed possible - not telling you who will win but the distance of advance, the amount of casualties, in detail. This can be backfitted to historical battles for validation. Granted it is expensive to get at the data (unless you have a peculiar sort of access - there are two ways in without paying the up front fee - which is astromical). Anyone who thinks they are incredably bad (or who does not know "he" is a "they") has not looked at this stuff. Even the most skeptical - and I probably qualify as one of those - would have to say "they did very well indeed."

They pleased their customers in the Pentagon, yes, but not the professionals in the field. When you're doing statistical modelling, you need at least three data points for each parameter you estimate. This includes interactions and higher order terms. They don't have the data to fit a model of that size. *And* given the importance of interactions in combat, their model is too small.

Actually there were a great many more than three. But the full data set costs (if memory serves) $10,000 to access - and the openly published material is just a sort of sampler. But one contractor has published the whole set - and I got a copy on CD rom. Even so, the Dupuy Institute continues to insist on the fee if you try to get even a little bit of the data.
Don't confuse what is in two books for the whole data set - or you will come to invalid conclusions.

At the time I was doing work in the area, they had a data set about one tenth the size of mine, and they had more free parameters than samples. Yet they published it! Nothing like putting your foot in it.